INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

BLOOD FLOW TO THE renal medulla has been shown to be altered independently of the blood flow to the renal cortex (2, 3). Interestingly, the sole blood supply to the renal medulla is through the descending vasa recta capillaries, which are derived entirely from the deep cortical postglomerular efferent arterioles (14). Descending vasa recta resemble capillaries because they are not surrounded by vascular smooth muscle cells but instead are surrounded by perivascular elements known as pericytes (10, 14). Pericytes are believed to be contractile in nature because they are found to possess the dense bodies and myofilaments that are necessary components of contraction (7). Recently, Pallone and associates have demonstrated that in vitro perfusion of rat outer medullary descending vasa recta is capable of vasomodulation when hormonally stimulated by angiotensin II (19) and adenosine (22). This indicates that these capillaries are capable of functional vasomodulation, at least at the level of the outer medulla, yet there is little information regarding the physical characteristics of the descending vasa recta pericytical myofilaments. The physiological function of pericytes could be of considerable consequence by acting to modulate blood flow within the renal medulla, a region not only important in the concentration of urine but also now shown to be important in the regulation of sodium excretion and the long-term control of arterial blood pressure (2, 3).

Recently, smooth muscle -actin has been found to be a useful marker for smooth muscle differentiation (23), and smooth muscle -actin has been proposed as a tool to distinguish pericytes from endothelial cells and fibroblasts (9) in the bovine retinal microcirculation. Smooth muscle -actin has been localized specifically to the vascular smooth muscle cells and precapillary pericytes in vivo in the rat renal cortex (1, 4) and rat mesangial cells in vitro (4). At present, it is not clear whether smooth muscle -actin is found within the renal medulla, specifically surrounding the descending vasa recta.

This study was therefore designed to determine the presence of smooth muscle -actin mRNA and protein within the pericytes of the renal medulla, using reverse transcription-polymerase chain reaction (RT-PCR) and fluorescent immunohistochemistry, respectively.

	METHODS

Top Abstract Introduction Methods Results Discussion References

Experimental animals. Tissue for the microdissection of the descending vasa recta and RT-PCR was obtained from adult, male Sprague-Dawley rats (200-275 g; Sasco, Madison, WI). Tissue from adult, male Sprague-Dawley rats (400-600 g) was used for the fluorescent immunohistochemistry. All animals were fed a standard pellet diet (Purina Mills, St. Louis, MO) and ad libitum water to drink. All protocols were approved by the Institutional Animal Care and Use Committee of the Medical College of Wisconsin.

Microdissection of outer medullary descending vasa recta. Outer medullary descending vasa recta (OMDVR) were microdissected as we have described previously (20). In brief, rats were first treated with furosemide (5 mg/kg ip). Thirty minutes later they were anesthetized with ketamine (50 mg/kg) and acepromazine (5 mg/kg) administered intramuscularly. A polyethylene catheter (PE-90) was inserted into the abdominal aorta distal to the left renal artery, and the left kidney was selectively perfused with 15-ml dissection solution prewarmed to 37°C followed by perfusion with 0.6 ml of 2.5% latex-coated brown-dyed microparticles (~1-4 µm in diameter; Polysciences, Warrington, PA). After the perfusion, the kidney was removed and cut coronally, and the renal medulla was excised. The medullary tissue was cut into smaller pieces and placed into the dissection solution containing 1 mg/ml collagenase (CLS 2, 192 U/mg; Worthington Biochemical, Freehold, NJ) at 37°C for 35-40 min. The kidney slices were rinsed with dissection solution and then placed under a Zeiss M3Z stereomicroscope (magnification ×16-100) for dissection. Four separate kidneys were used in the microdissection of the OMDVR, in which each kidney supplied ~20-25 individual OMDVR from the inner stripe of several outer medullary vasa recta bundles. Vessel lengths were measured by an optical micrometer, and ~12 mm of OMDVR were isolated per kidney dissection. The dissected vessels were transferred to another dish to wash off any contaminating debris and then transferred to a thin-walled ultracentrifuge tube containing 100 µl TRIZOL reagent (GIBCO-BRL, Gaithersburg, MD) for permeabilization of the isolated tissue.

RNA isolation and RT. Total RNA was extracted from the isolated OMDVR (~12 mm in total length) as previously described by adding 50 µl chloroform. The aqueous phase was removed after centrifugation, and the RNA was precipitated with isopropanol (70 µl). The RNA pellet was then washed with 75% ethanol, allowed to dry at room temperature, and resuspended in deoxyribonuclease (DNase) solution containing 1 U RQ1 ribonuclease-free DNase and 20 U RNasin (Promega Biotech, Madison, WI) to remove any genomic DNA contamination. The RNA was then reextracted as described above, and the RNA pellet was allowed to dry at room temperature for 5 min. RT was performed on the total RNA using 0.5 µg oligo(dT)₁₅₁₈ (Promega Biotech) as previously described (19). cDNA was then synthesized at 37°C for 60 min, and the reaction was stopped by heating to 95°C for 5 min.

Preparation of oligonucleotide primers. All nucleotide primers were purchased from Operon Technologies (Alameda, CA). Oligonucleotide primers are shown in Table 1 and were chosen from the published cDNA sequences of rat smooth muscle -actin (12), human -actin (7), and the vasopressin V₂ receptor (V₂R) (10). The primers for -actin and smooth muscle -actin primers were designed to span at least one intron. The V₂R primers are found on the same exon.

PCR. All PCR reactions were performed in a total volume of 50 µl in the presence of (in mM) 0.2 dNTP, 10 1,4-dithiothreitol, 50 KCl, 1.0 MgCl₂, and 10 tris(hydroxymethyl)aminomethane (Tris) · HCl, pH 8.3, as well as 50 pmol of each primer, 2.5 U AmpliTaq polymerase (Perkin-Elmer Cetus, Norwalk, CT). Mineral oil was layered on top of each sample to prevent evaporation of the liquid. All primer cDNA amplifications were optimum under these standard conditions. Seven-microliter aliquots of the RT reactions were used in the amplification for both smooth muscle -actin and V₂R, whereas 1 µl was used for -actin. The reaction mixture was first denatured at 94°C for 5 min and then cycled for 35 cycles between 94°C (denaturation) for 1 min, 64°C (annealing) for 1 min, and 72°C (extension) for 1 min. Samples were incubated for an additional 7 min at 72°C after the completion of the final cycle.

PCR product analysis. From each of the PCR reactions, 10-µl aliquots were size-fractionated by electrophoresis on a 1.6% agarose gel. After electrophoresis and ethidium bromide staining, DNA bands were visualized with an ultraviolet transilluminator. To verify the authenticity of the PCR products, the gels were denatured, neutralized, and blotted onto a nylon membrane (Micron Separations) for Southern blot analysis. The DNA was cross-linked to the membrane with ultraviolet light and hybridized with an internal oligonucleotide, which was labeled 3' with a fluorescein nucleotide (Amersham UK). The final blot wash was 1× saline-sodium citrate-0.1% sodium dodecyl sulfate (SDS) at 58°C. To further authenticate the specificity of the PCR products, PCR sequencing using the dideoxynucleotide method was performed (Amersham, Arlington Heights, IL).

Protein isolation. The kidneys from anesthetized male Sprague-Dawley rats were cut in a coronal section to isolate the outer medulla. The outer medullary tissue was homogenized in (in mM) 5 K₂HPO₄, 5 KH₂PO₄, 250 sucrose, pH 7.7, 0.1 EDTA, 0.1 phenylmethylsulfonyl fluoride, 2 µg/µl leupeptin, and 5 µg/µl pepstatin. The homogenate was centrifuged at 1,000 g for 10 min to remove any incompletely homogenized membrane fragments and nuclei, and the supernatant was centrifuged at 16,000 g for 20 min. The 16,000-g supernatant was removed for determination of protein concentration using the Coomassie method (Pierce, Rockford, IL), and the proteins were frozen at 80°C.

Electrophoresis and immunoblotting of membranes. Sample buffer [2% SDS, 100 mM Tris · HCl, pH 6.8, 5% -mercaptoethanol, 12% (vol/vol) glycerol, and 0.02% (wt/vol) Bromphenol blue] was added to the protein sample, and the mixture was heated to 100°C for 5 min. Five micrograms of outer medullary protein were loaded onto the gel and then size separated by electrophoresis through a 12% SDS-polyacrylamide gel. The proteins were transferred onto a nitrocellulose membrane (Bio-Rad), and the membranes were blocked with 15% nonfat dried milk in blotting solution overnight at 4°C. The blotting solution contained (in mM) 137 NaCl, 20 Tris · HCl, pH 7.4, and 0.08% Tween 20. The membranes were incubated for 30 min with the monoclonal smooth muscle -actin primary antibody (1:2,500; Sigma, St. Louis, MO) at room temperature. The membranes were washed in several changes of blotting buffer, and then incubated for 30 min with secondary antibody (goat anti-rabbit immunoglobulin G, 1:1,000; Bio-Rad). The membranes were washed, and the protein bands were detected by chemiluminescence (WesternView, Transduction Laboratories) on X-ray film.

Immunohistochemistry of the renal outer and inner medulla. Kidneys were perfusion-cleared with Tyrode medium and subsequently perfused with india ink in Tyrode medium. The kidney tissue was hemisected and cut into wedges containing the cortex and outer and inner medulla. The wedges were cryoprotected by treatment with increasing concentrations of sucrose (10, 20, and 30%) in phosphate-buffered saline (PBS), pH 7.4. The tissue was then frozen on dry ice in OCT embedding medium (Scientific Products, McGaw Park, IL) and stored in liquid nitrogen. Frozen sections (50 µm thick) were made using a cryostat set at 20°C, and sections were placed on glass slides ringed with rubber cement. The sections were thawed for 5 min, fixed for 10 min in 95% ethanol, rinsed in PBS, and then rinsed in PBS containing Triton X-100 (Sigma). They were incubated with the smooth muscle -actin antibody (clone 1A4; Sigma) at a concentration of 1:50 in PBS at room temperature for 1 h. The sections were washed in PBS containing bovine serum albumin and were subsequently incubated in secondary antibody (goat anti-mouse labeled with rhodamine, 1:50 in PBS; Boehringer-Mannheim) for 1 h at room temperature. The sections were washed in PBS and then mounted in 50% glycerol containing 1% p-phenylenediamine (PPD). The sections were studied with a Nikon Optiphot fluorescence microscope.

Kidneys that were used for antibody staining of 0.5- to 1-mm-thick slices were cleared as described above and fixed by perfusion of 4% ammonium molybdate in distilled water. The kidney was hemisected and cut into wedges as above, hand sliced, and then processed as described for cryosectioned tissue. The thick slices were mounted in PPD on large glass slides for study.

	RESULTS

Top Abstract Introduction Methods Results Discussion References

Smooth muscle -actin mRNA expression in OMDVR. OMDVR from the outer medullary vasa recta bundles were microdissected and total RNA was extracted to determine whether these isolated vessels expressed smooth muscle -actin mRNA. In each experiment, a pool of descending vasa recta totaling ~12 mm in length was obtained by microdissection from each individual kidney. OMDVR were individually microdissected and inspected under ×100 magnification to achieve the purest microvessel preparation. Descending vasa recta were identified by their very thin (<20 µm in diameter) and "bumpy" wall appearance, which was related to the presence of the cell bodies of the pericyte. Figure 1 illustrates the clean microdissection of an isolated OMDVR from the vascular bundles with no evident contaminating debris from surrounding structures, particularly renal tubules. This lack of debris is an important and stringent criterion, which was implemented to accept these vessels for RNA extraction and RT-PCR. Figure 2 shows the RT-PCR results of the total RNA from isolated OMDVR for smooth muscle -actin (Fig. 2A), -actin (Fig. 2C), and V₂R (Fig. 2D). In each microdissection and RT-PCR that was performed (n = 4), outer medullary tissue RNA was used as a positive control (lane 2) and lane 3 shows that microdissected OMDVR express smooth muscle -actin. In all of the kidneys used for microdissection, the presence of smooth muscle -actin was clearly evident (n = 4). The specificity of the RT-PCR products were shown with Southern blot analysis using an internal oligonucleotide probe for smooth muscle -actin (Fig. 2B) and by subsequent sequencing of the RT-PCR products (which showed that the PCR product was 100% homologous compared with smooth muscle -actin sequences found within GenBank).

View larger version (76K): [in this window] [in a new window]   Fig. 1.   Micrograph of a microdissected outer medullary descending vasa recta capillary from a vascular bundle (magnification ×200).

View larger version (56K): [in this window] [in a new window]   Fig. 2.   Ethidium bromide-stained agarose gels of reverse transcription-polymerase chain reaction (RT-PCR) products from outer medullary tissue (lane 2; positive control) and microdissected outer medullary descending vasa recta (lane 3) for smooth muscle -actin (A), -actin (C), and vasopressin V2 receptor (D). Corresponding Southern blot analyses for smooth muscle -actin (B) verify the specificity of the amplified RT-PCR products. Expected size of the PCR product was 389 bp for smooth muscle -actin, 461 bp for the V2 receptor, and 350 bp for -actin; the positions of the DNA size markers (in bp) are shown at left. Negative control samples are shown in lane 4 (PCR amplification of sterile water), lane 5 (PCR amplification of dissection solution), and lane 6 (PCR amplification of DNase-treated RNA).

RT-PCR of the V₂R mRNA, which we have described previously (19), was used as a negative control to ascertain tubular contamination (i.e., medullary thick ascending limb of Henle and/or outer medullary collecting ducts), which can occur during the microdissection and RNA extraction of OMDVR. The absence of V₂R mRNA within the OMDVR (Fig. 2D, lane 3) verified the purity of these microvessels and lack of tubular contamination.

The results from RT-PCR demonstrate that the mRNA for smooth muscle -actin is found within microdissected OMDVR (lane 3). Bands were not detected in any of the negative control samples, which included, in lane 4, PCR amplification of sterile water; in lane 5, PCR amplification of dissection solution; and in lane 6, PCR amplification of DNase-treated RNA (V₂R). The negative control for the PCR amplification of DNase-treated RNA was necessary for only the V₂R primers because these primers did not span introns like those for smooth muscle -actin and -actin.

Specificity of smooth muscle -actin antibody by immunoblot analysis using outer medullary protein. Initial experiments were performed to determine the specificity of the monoclonal smooth muscle -actin antibody in protein preparations from the outer medulla. Figure 3 shows that the monoclonal antibody specifically recognizes a 47-kDa protein, which is the expected size of smooth muscle -actin, in outer medullary protein isolated from two different rat outer medullas (lanes 1 and 2).

View larger version (85K): [in this window] [in a new window]   Fig. 3.   Specificity of smooth muscle -actin was shown by Western blot analysis using outer medullary protein (5 µg/lane; lanes 1 and 2). Lanes 1 and 2 are protein isolated from two different rat outer medullas (OM). Protein standards are indicated at left. Molecular weight of smooth muscle -actin is 47 kDa (as indicated by arrow on right).

Fluorescent immunohistochemistry of the outer and inner medulla. The specificity of the monoclonal antibody allowed for the immunolocalization of smooth muscle -actin within sections of the outer and inner medulla of the kidney. Figure 4 demonstrates the fluorescence of smooth muscle -actin surrounding descending vasa recta from the outer (Fig. 4, A-D) and inner medulla (Fig. 4E) of hand-sliced tissue. Figure 4A (magnification ×200) and Fig. 4B (magnification ×1,000) show an efferent arteriole branching into a vasa recta bundle near the outer-inner stripe of the outer medulla. As shown by the small arrow, the immunofluorescence was localized to the cell body of the pericyte and its extensions. Figure 4, C and D, demonstrates the smooth muscle -actin within descending vasa recta in the inner stripe of the outer medulla. Figure 4E shows inner medullary descending vasa recta also exhibiting abundant smooth muscle -actin.

View larger version (134K): [in this window] [in a new window]   View larger version (87K): [in this window] [in a new window]   Fig. 4.   Fluorescently labeled anti-smooth muscle -actin-stained pericytes along the proximal vasa recta bundle near the outer-inner stripe border of the outer medulla (A and B) in hand-sliced tissue. Arrows in A (magnification ×200) and B (magnification ×1,000) point to the same pericytic cell body surrounding a descending vasa recta capillary. C and D show smooth muscle -actin immunofluorescence of descending vasa recta from different vasa recta bundles in the inner stripe of the outer medulla. E shows immunofluorescence for smooth muscle -actin on inner medullary descending vasa recta. C-E are at ×1,000 magnification. Solid bar = 60 µm in A and 12 µm in B-E.

Interestingly, as shown in Fig. 5, which was from a cryostat section of renal medullary tissue, the immunofluorescence of smooth muscle -actin was found to continue deep within the inner medulla and disappeared only near the tip of the papilla. Because this kidney was only partially filled with india ink, a better visualization of several outer medullary vascular bundles was achieved. Smooth muscle -actin fluorescence was clearly present in the pericytic cell bodies and its tentacle-like extensions, which surrounded the descending vasa recta. Because of the specificity of the smooth muscle -actin antibody in the pericytes, descending vasa recta could be immunofluorescently distinguished from other vascular, tubular, and interstitial structures. Moreover, it is important to note that ascending vasa recta are not associated with pericytes, and thus no immunofluorescence was observed. No immunofluorescence was detected in about the distal one-third of the inner medulla. It is evident from Figs. 4 and 5 that the pericytes surrounding the outer and inner medullary vasa recta capillaries are circumferentially well positioned to alter microvascular diameters when stimulated to contract.

View larger version (154K): [in this window] [in a new window]   Fig. 5.   Immunofluorescence for smooth muscle -actin (left) and light microscopy (right) of inner medullary tissue intermittently filled with india ink in cryostat-sectioned tissue. Note the specific immunofluorescence of the descending vasa recta from the inner stripe of the outer medulla down to the inner medulla. No immunofluorescence was observed in ascending vasa recta. Black arrow shows approximate border between the inner stripe of the outer medulla and the inner medulla. Specific fluorescence for smooth muscle -actin descended to near the papillary tip of the kidney section. Magnification ×100; solid bar = 120 µm.

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

The application of molecular biological techniques has allowed for the determination of gene transcription and subsequent protein translation within minute tissue samples such as microdissected tubules and blood vessels as was required for the present studies. RT-PCR is an extremely sensitive technique that allows for the amplification of reverse-transcribed cDNA from nearly single copies of mRNA transcripts, but it does not provide information regarding the cellular origin of the mRNA. Moreover, the presence of the mRNA does not assure that protein is being translated from the mRNA within the specific cells of interest. For this reason, in the present study, fluorescent immunohistochemistry using antibodies specific for smooth muscle -actin was used to localize specific cell type(s) that translated the smooth muscle -actin mRNA into proteins. By coupling RT-PCR with immunohistochemistry, the results of this study determined the localization of the renal medullary gene expression of smooth muscle -actin and further discriminated that the cellular site of protein translation was within the vasa recta pericytes.

The importance of contractile smooth muscle -actin found within renal medullary pericytes relates to the role of these cells in determining the vascular diameter of the vasa recta and consequently the regulation of renal medullary blood flow. Contractile elements, specifically smooth muscle -actin, have been found within pericytes that surround the extrarenal microcirculation of the bovine retina and rat mesentery (9, 18). In the kidney, there is definitive evidence that vascular smooth muscle cells in the renal cortex of rats possess smooth muscle -actin (1, 4), but there is little known about its existence within the renal medulla. Histological studies by Moffatt (14) demonstrated that renal medullary pericytes surrounded descending vasa recta and that these pericytes contained myofibrils that were similar to those found within the vascular smooth muscle cells, but the type of myofibril was not determined. Carey et al. (1) determined the expression of smooth muscle -actin within the renal medulla of Wistar rats but concluded that its existence was developmentally regulated because these investigators observed smooth muscle -actin within the renal medulla in very young (15 day old and younger) rats. The results of the present study conclusively demonstrate the presence of smooth muscle -actin in the pericytes of medullary descending vasa recta in adult rats. The divergence in the results may be attributed to strain differences (Wistar vs. Sprague Dawley) but is more likely due to differences in the methodological approaches used in the detection of smooth muscle -actin (fluorescence vs. diaminobenzidine staining).

Pericytes within the kidney are heterogeneous in nature (21) and may participate in several ways to modify microvessel function. Morphologically, pericytes of the peritubular capillaries in the renal cortex are aligned along the longitudinal axis of the vessels (4), whereas the pericytes surrounding the descending vasa recta in the renal medulla are aligned longitudinally and circumferentially to the vessels as seen in Fig. 4. Consistent with these morphological characteristics, Murphy and Wagner (15) found that cultured pericytes were capable of contracting in two ways: first, tangentially (longitudinally) to the vessel, which would be expected to increase the permeability of the capillaries, or, second, circumferentially, which would modulate vessel diameter and modulate blood flow. This would be consistent with observations by Pallone and associates (19, 22), who have seen reductions of descending vasa recta diameters in isolated perfused vasa recta capillaries from the rat outer medulla superfused with hormonal vasoconstrictors.

A potentially important physiological role for the presence of smooth muscle -actin within inner medullary descending vasa recta is related to the ability of these contractile elements to differentially regulate blood flow distribution within the renal medulla. In vivo functional studies in our laboratory (12, 16, 17) using implanted laser-Doppler flowmetry techniques have now shown that selective infusion of vasoactive compounds into the renal medullary interstitial space can lead to preferential modulation of renal medullary blood flow. Franchini and Cowley (6) have also observed that 48-h water restriction in conscious Sprague-Dawley rats preferentially reduced inner medullary blood flow without altering the blood flow to the outer medulla. In addition, videomicroscopic studies by Fenoy and Roman (5) showed that volume expansion with intravenous saline infusion was associated with an increase in the number of functionally perfused vasa recta capillaries within the renal medulla, suggesting a mechanism for recruitment of previously unperfused descending vasa recta. The observations from these studies and the visually observed reductions of OMDVR diameter with angiotensin and vasopressin (19, 24) indicate that the pericytes play a pivotal role in the modulation of renal medullary blood flow.

In conclusion, this study has shown that smooth muscle -actin mRNA is found in microdissected descending vasa recta capillaries and that the protein is translated and found specifically within the pericytes surrounding these vasa recta capillaries within the outer and inner medulla.